this project stores millions of text chunks inside a video file (mp4)
then runs sub-second semantic search on it
- no vector DB, no servers
- uses 10x less RAM & storage
- no internet required
it's called Memvid and it just broke my brain
Step 1: Raise $15M
Step 2: Burn it to generate $8.3M
Step 3: Multiply by 12 and call it $100M ARR
Step 4: Tweet it
Step 5: Raise $200M
Still no users
Still no profit
Welcome to the new economy
You did the impossible:
Built a new browser (Arc)
Actually made it better than Chrome
Everyone loved it, devs, influencers, tech Twitter, YouTube
Hype off the charts
$50M raised at a $550M valuation
You cracked the hardest problems in startups:
โ attention, distribution,
i dont think you understand
you catch 3am dopamine and vibe code a saas
you drop $49 on a .ai domain
you launch on producthunt, tweet it, it gets 7 likes
you get 2 signups, $0 MRR, gmail accounts, one asks if itโs free
your net margins are -200% from vercel and clerk auth
Need an app idea?
1. Open Ahrefs
2. Type in a domain with AI tools
3. View organic keywords
4. Filter on Keyword: Alternative
5. Add CPC > 1.00 to filter for keywords in demand
You'll see a list of tools people are NOT happy with!
Fun fact: I removed Axios and I got +200% better latency.
They have some bug causing slowdowns when sending auth headers.
Literally went from 500ms to 3ms ๐คฏ
โ ๏ธ PSA for Cursor AI users that...
- install random .cursorrules from directories
- enable yolo mode in cursor settings
- connect to MCP servers you found on discord
- use web search to and get prompt injected by @elder_plinius
This means @cursor_ai can literally run
- rm -rf
Just set up a custom ESLint rule to enforce 200-line max file size and it's a game changer:
- Forces splitting up code into files
- Makes @cursor_ai more effective in getting context
- To fix, just click "Fix in Chat"
Clean code is best code ๐ Wish I'd done this years ago!
๐ญ AI has completely broken my ability to code
I can't code manually anymore (feels slow knowing Ai exists)
I can't use AI either (results feel janky, hard to trust)
So I just feel paralyzed, nauseous, stuck, thinking of doing an AI coding detox
Anyone else notice this?
